Comments on the proposed revisions to the 1929 Hurricane Season
From the Best Track Supervisory Committee
[Reponses to the Committee are bold faced and in brackets…July 2010]
Storm #1, 1929:
1. Regarding the revised track, a kink has been introduced around 0000 UTC June 28. Can this be smoothed out? Surely the precision in the revised best track locations does not warrant such a discontinuity.
[Agreed, the kink has been smoothed.]
2. The Monthly Weather Review article for this system states that the swath of high winds was 16 to 20 miles across. Given this, is it possible that the 13 n mi RMW used here is too large? If so, does that change the landfall pressure and intensity?
[The small RMW of 13 nm estimated by Ho et al. appears to be somewhat too large given the swath of high winds 16 to 20 miles across reported in the Monthly Weather Review. The RMW is analyzed to be about 10 nm, which would be quite small compared with 22 nm climatologically for hurricanes at the same latitude and central pressure – Vickery et al. 2000. Given this, the small outer closed isobar (150 nmi radius), somewhat low environmental pressures (1007 mb outer closed isobar), a value stronger than that suggested by the pressure-wind relationship is chosen: 80 kt is the estimated maximum sustained winds at landfall. Winds are thus boosted at 18Z on the 28th from 70 to 80 kt. This retains the system as a Category 1 hurricane for central Texas (“BTX1”), in agreement with the original HURDAT. This is now indicated in the writeup.]
3. Are there any OMR records available for Victoria or Port O’Connor that would provide additional information?
[No. These were not official Weather Bureau stations, so no OMR hourly data are available.]
4. The committee concurs with the change from extratropical transition to dissipation. You might want to discuss this in more detail.
[Originally, HURDAT had this system becoming extratropical around 06Z on the 29th. However, analysis of the synoptic environment show no significant baroclinic zones (or separate extratropical cyclones) and the cyclone’s structure remained fairly symmetric. These indicate that instead of a transition to an extratropical cyclone, it remained a tropical cyclone until dissipation. This is now indicated in the writeup.]
5. The Severe Local Storms table in the June Monthly Weather Review (pg. 265) mentions wind damage in El Paso, Texas around 8 PM June 29. Could this be related to this hurricane? Extrapolating the track from the last point puts the system or its remnants not far from El Paso at that time.
[Obtaining the Original Monthly Records for El Paso from NCDC’s EDADS site reveals that the TC did indeed make a rare strike on El Paso as a tropical storm. The winds were 32 kt (5 min winds) SW at 22Z on the 29th; 7 kt (hourly winds, no 5 min winds available) S at 23Z; 3 kt (hourly) N with 1008 mb pressure, 79F temperature, 58F dewpoint (highest for the month), and .0.45” rainfall (last 12 hr) at 00Z on the 30th, and 40 kt (5 min) NE at 01Z. Commentary in OMR for El Paso stated: “The only appreciable precipitation during the month occurred during thundershowers on the 29th. Excessive precipitation occurred at that time…Strong winds also accompanied the thunderstorm and did some damage to roofs, plate glass windows, etc.” Best track is now extended through 06Z on the 30th with 40 kt tropical storm intensity indicated at 00Z on the 30th.]
6. Metadata p.2, last paragraph, line 3, replace “underweant” with “underwent”. Use spell checker on entire document.
[Corrected. Spell checker done on entire document now with several corrections added in.]
Storm #2, 1929:
1. Although it is likely that the genesis of this hurricane occurred earlier than September 22, it is dubious that the cyclone existed as a 30-kt depression for nearly 3 days prior to becoming a tropical storm. Do the data make a strong enough case for this system to have developed as early as September 19?
[The observations from the 19th to the 21st are quite definitive in demonstrating that a closed, non-baroclinic low was present. Indeed, one could even make the case to extend genesis back to the 18th as well. While it is somewhat unusual for a system to remain as a tropical depression for three days after genesis, observations are not available to indicate/confirm that the system was a stronger tropical cyclone. Without evidence for such a status, the system likely has to remain classified as a tropical depression from the 19th until the 21st.]
2. The MWR account says that the vessel Potomac measured a pressure of 924 mb on September 25 about 15 miles west of Abaco Island. Given the track of the hurricane, this is not possible. The metadata should indicate that the reference to Abaco is undoubtedly an error, and that the island in question is probably Eleuthera. Please correct this.
3. Several questions about the Florida Keys landfall: First, did the center pass over Key Largo or over Long Key, which is several miles to the southwest? Both are referenced as landfall sites in the September 28 text. Please note that the 10-minute lull mentioned for Key Largo is not consistent with a direct hit from a slow-moving 28 n mi RMW, suggesting that it was closer to the edge of the eye than it was to the center. Second, where is the 948 mb observation in Key Largo actually mentioned in the metadata? There are several second-hand references, but the original source of the ob never seems to be referenced. What is the original source of this ob? Does it actually exist, or is it someone’s estimate that has been incorrectly assumed to be an ob?
[The best estimate is that the center of the hurricane made landfall between Key Largo and Long Key near 25.0N 80.5W – over Plantation Key. It is agreed that the 10-min lull in Key Largo was in the edge of the eye, rather than the center. After further investigation, the 948 mb reading – a very suspiciously exact 28.00” – for Key Largo was likely an estimate, not a measurement. There is no mention of this value in either the Key West OMR, the Climatological Data – Florida Section, or the Monthly Weather Review. The first time this value appears in Dunn and Miller’s 1960 textbook. It has since been repeated in Ho et al., Jarrell et al., and Barnes. However, given the observed reading of 954 mb at 1330Z at Long Key on the 28th which was likely in the edge of the eye, a central pressure of 948 mb is quite reasonable and is retained in HURDAT. Track slightly altered to better accommodate a 13Z landfall over Plantation Key.]
4. Was the 975 mb pressure at Panama City actually a central pressure as mentioned in the summary section? If so, that observation -- along with the 5-minute wind of 61 kt at Pensacola, from an offshore direction, on the weak side and well away from the center suggests a stronger than Category 1 hurricane at landfall in the Florida Panhandle. Indeed, the original HURDAT showed Category 2. An upward revision of the Panhandle landfall intensity should be considered.
[The 975 mb pressure measurement at Panama City was analyzed by Ho et al. to be a central pressure reading. However, there is no definitive evidence that this is the case. Given that the original track kept the center of the hurricane just east of the town, the revised track will as well and treat the 975 mb value as a peripheral pressure reading. Landfall is kept as an 85 kt Category 2 hurricane
5. Does a detailed analysis support moving the time of extratropical transition forward? This is a somewhat unclimatological transition made more so by moving it forward 12 hours.
[The transition to extratropical is supported by observations and analyses from the Historical Weather Map series supplemented by COADS ship data and the Original Monthly Records from the U.S. Weather Bureau stations. These show fairly distinctly that the system became extratropical by 12Z on the 1st of October, soon after landfall in the Panhandle of Florida. Note the significant (20F) north-south temperature gradient through the cyclone and asymmetric structure of the winds and pressure field – all of which supports extratropical transition by 12Z on the 1st, if not earlier.]
6. What significant land observations of the extratropical low are available as it crosses the eastern United States? The metadata descriptions for October 2-5 mention only the marine obs.
[Peak tropical storm force winds have now been included both in the daily summary as well as the spreadsheet.]
Storm #3, 1929:
1. The committee concurs with the addition of this system. There is a concern about there not being enough evidence that it was actually tropical since the Historical Weather Maps kept a front attached to it through its lifetime. If possible, please provide a more detailed analysis.
[Synoptic analyses independent of the Historical Weather Maps charts have been performed and confirm a tropical cyclone (closed low, non-baroclinic) of tropical storm intensity was present on the 25th and 26th of September.]
Storm #4, 1929:
1. The committee concurs with separating this storm into two different storms. However, the analyses presented in the binder do not do a good job in showing the two different systems existing nearly simultaneously. Please provide a data plot at the time where both circulations were most obviously present.
[Data for 12Z (utilizing obs from 06 to 18Z) on the 19th of October are re-plotted, which does provide substantial evidence for both systems existing at the same time. There are insufficient observations at 00Z on the 20th of October for a meaningful synoptic assessment.]
2. There are two ship reports of 60 kt from this system, one on 15 October and the other on 19 October. Both of them are about 60 n mi from the proposed tracks, which brings up the possibility they were both outside the RMW and didn’t sample the maximum winds. Please re-examine the data to see if this system reached hurricane strength.
[For the 60 kt report on the 15th, the ship – which reported every four hours on that date while in the cyclone – only reported their position rounded to a precision of 1.0 degree latitude and 1.0 degree longitude (e.g., at 03Z they were at 33.5N 38.5W, at 07Z 32.5N 38.5W, at 11Z 32.5N 39.5W, etc.). This means that the ship position has an uncertainty of +/- 30 nm both in the north-south and in the east-west positions. Thus it is unknown as to the exact location of the ship relative to the cyclone, though the best guess is about 60 nm (+/- 45 nm). Because of this, it is also unsure whether the ship was near the RMW at the time it measured the 60 kt. The ship on the 19th which reported the 60 kt wind does appear to be reporting its position to the nearest 0.1 degree latitude and longitude, but only one measurement was reported on that date. The cyclone on the 19th was undergoing extratropical transition on that date and it may very well be the case that the 60 kt was observed near the RMW in the right front (strong) quadrant of the system. Thus 60 kt is retained as the peak intensity for this system, though it is quite possible that this system reached hurricane intensity at some point during its lifetime.]
Storm #5, 1929:
1. The committee concurs with this addition, with the comment above that a data plot which shows the two separate circulations existing simultaneously would be useful
[See reply above.]
System #1: The committee concurs with leaving this system out of HURDAT.
System #2: The committee concurs with leaving this system out of HURDAT. Note that there are no charts on this system in the 1929 binder.
System #3: The committee concurs with leaving this system out of HURDAT. Are there any observations close enough to estimate a central pressure?
System #4: The committee concurs with leaving this system out of HURDAT, with three comments. First, it looks like there is a very strong pressure gradient on the north side of the low, which suggests the possibility of gales even if none were observed. Second, that strong pressure gradient was accompanied by temperatures of 64 F north of the low on the 23 September map, which brings of the possibility of frontal involvement. Third, the 25 September map shows a strong vorticity or circulation center just southeast of the mouth of the Mississippi River? What is this feature and how does it relate to depression? These should be noted in the write-up.
[All of these comments are now included and elaborated on in the Additional Notes section.]
System #5: The committee concurs with leaving this system out of HURDAT.
System #6: The committee notes that the central pressure of this system seems to be low enough to support tropical storm winds. What land station data is available from Cuba and South Florida as the system passed? It is likely this system was not a tropical storm, but the committee would like a second look and more detail in the write-up.
[The Climatological Data for Florida was obtained for October 1929 to better examine this system. The US Weather Bureau had six official stations, the closest to this system being Key West and Miami. Neither reported tropical storm force winds, though Key West did reach a minimum sea level pressure for the month of 1005 mb on the 21st. Such a low pressure would often be accompanied by tropical storm intensity winds. However, as the system was in the warm sector of a large, developing extratropical system, the environmental pressures were quite low (roughly 1007 mb outer closed isobar). There were also a few dozen cooperative observing stations in Florida as well, but none of these had barometric pressure or wind observation readings recorded. No significant impacts from this system were noted in the Climatological Data and it is likely that the system did not reach tropical storm intensity over Florida. Likewise, Ramon Perez of the Cuban Climate Institute indicated that this system was also of tropical depression intensity while crossing Cuba early on the 21st. Thus the system is not included in HURDAT.]
System #7: The committee concurs with leaving this system out of HURDAT. Note there was a 30 mph ship with 1008 mb well to the northeast of the proposed position on 2 November, suggesting either a very large system or a more northeasterly position on that day.
Other comments: The Monthly Weather Review indicates three systems that should be investigated to see if they were tropical cyclones, or for possible inclusion in the Additional notes section.
1. The July Monthly Weather Review mentions a low pressure area that caused a small-scale heavy rain event at Cape Hatteras on 21 July. The July Cyclone Tracks Maps shows the low came up to the Hatteras area from the south. There are no obvious indications that this was a tropical storm, but it should be investigated to see if it was a depression.
[The Historical Weather Map series shows that this system was a short-lived frontal low without any tropical storm force winds. It has now been added into the Additional Notes section.]
2. The August Monthly Weather Review mentions a tropical disturbance of “limited extent and intensity” that was near St. Lucia on 19 August and northwest of Grand Cayman on 22 August. While this sounds like a tropical wave, it is probably worth a look to see if it was something more.
[The Historical Weather Map series shows that this system was a moderate tropical wave moving across the Caribbean. However, on the four days that the system was evident, there was no evidence of a closed low. This system has now been added into the Additional Notes section.]
3. The August Monthly Weather Review Cyclone Tracks Map shows a low that formed in the Gulf of Mexico on 27 August and moved northeast across Florida into the Atlantic. There is no other mention of this system, which suggests it was not very strong. However, it may also be worth a closer look.
[The Historical Weather Map series shows that this system was an open trough on the 27th and 28th of August. It did form a closed low on the 29th when it was off of the Georgia coast, but the system by then was embedded in a frontal boundary and considered extratropical. No gales were observed from the 27th through the 29th. The system has now been added into the Additional Notes section.]